Subscribe to DTC Newsletter - ⁠https://dtcnews.link/signup⁠In this third appearance on the DTC Podcast, Constantine Yurevich, founder of SegmentStream, delivers the most comprehensive breakdown of the marketing measurement landscape we've ever recorded.This isn’t theory. It’s a raw, truth-filled look at what works, what’s broken, and how DTC brands can actually understand what’s driving growth.Forget the vendor noise. This episode will reset your measurement mindset.Visit SegmentStream website: https://segmentstream.com/ Follow Constantine Yurevich on LinkedIn: https://www.linkedin.com/in/yurevichcv/Key Insights and Takeaways:The 3 measurement models in DTC: empirical attribution, statistical modeling (MMM), and causal testing (geo-holdouts)—and why only one is truly useful for most brands.Why MMM is mostly theater: assumptions-driven, biased by design, and rarely accurate for DTC brands under $100M in spend.The myth of incrementality measurement: how geo-holdouts and lift tests produce confidence intervals too wide to act on.How to build a trustworthy attribution stack:First-click attribution (and why it’s ignored by platforms)Identity graph stitching across devices/usersSelf-reported attribution + re-attribution logic to uncover dark-funnel sources (YouTube, podcasts, etc.)Why first-click attribution is the best tool for understanding new customer acquisition.How self-reporting + re-attribution can show you where brand awareness is really coming from.The fatal flaw in measuring “incrementality” via platform lift studies.How to use marginal analytics to reallocate budget before returns diminish.Why platforms are killing first-click visibility—and what that tells you about their incentives.This episode is for DTC operators who are tired of overpaying for last-click conversions that would’ve happened anyway, want clear signals from upper-funnel and brand channels, and are hungry for real insight.Timestamps00:00 Intro and the state of marketing measurement02:15 Why most models are disconnected from reality05:00 Three main categories of marketing measurement08:10 The limits of MMM and mathematical models12:00 The rise of new-gen MMM and confirmation bias17:00 Geo holdout testing and its flaws24:00 Why true incrementality is nearly impossible to measure31:00 First click vs last click attribution37:00 Using identity graphs and self-reported attribution44:00 Why attribution still matters after 20 years52:00 Where SegmentStream fits in and key takeawaysHashtags#DTC #Ecommerce #MarketingMeasurement #Attribution #MMM #DigitalMarketing #BrandGrowth #MarketingData #Incrementality #AdTechSubscribe to DTC Newsletter - https://dtcnews.link/signup

Advertise on DTC - https://dtcnews.link/advertise

Work with Pilothouse - https://dtcnews.link/pilothouse

Follow us on Instagram & Twitter - @dtcnewsletter

Watch this interview on YouTube - https://dtcnews.link/video